Professional Documents
Culture Documents
By
© April, 2012
CERTIFICATION
1
I hereby declare that this submission is my own work towards the M.Sc. and that, to the best of
which has been accepted for the award of any other degree of the University, except where due
Certified by:
DR. S. K. AMPONSAH
Certified by:
Certified by:
DEDICATION
2
To my mother, my wife: Saviour Adzimani and all my siblings for their support.
3
ACKNOWLEDGEMENT
The completion of this research represents the realization of a long awaited dream.
My deepest heartfelt thanks: first of all goes to the Almighty God for granting me guidance and
I express my profound gratitude to my supervisor and mentor, Dr. S. K. Amponsah from the
proposal stage to the completion stage. All these thoughtful comments were considered carefully
in the final preparation of the project and they have been of invaluable help in reinforcing my
I am also very grateful to all my lecturers in the Mathematics and Statistics Department, Kwame
Nkrumah University of Science and Technology (KNUST) for being helpful and impacting
I shall forever remain indebted to the staff of the Ghana Highways Authority for providing me
Last but not the least, I acknowledge the following people: Nixson K. Gborgenu, Saviour
Adzimani and Ohene Mamucan. Ohene Mamucan helped me in typing the work and Nixson and
Saviour gave me their spiritual and moral support through the entire period of this programme.
4
ABSTRACT
The demand for the use of natural gas is on the increase as an energy source. Natural gas
transportation requires a continuous pipeline network from the source of gas across long distance
to the various destinations. The main objective involves extending gas pipelines from Takoradi
to all the regional capital towns in Ghana by developing a straight forward method of locating
pipeline facilities and designing pipeline networks to minimize cost and distance covered. The
problem is formulated as a network of distances and the solution is presented based on Prim‟s
Algorithm for minimum connections. Data on distances are obtained from the Ghana Highways
Authority. In comparison, the total distance in the original network was reduced drastically by
68.79% to the new network. Solutions are provided for the environmental damages revealed by
5
TABLE OF CONTENTS
Title Page
CERTIFICATION………………………..…….…………………………………………………ii
DEDICATION…………………………………………………………………………………….ii
ACKNOWLEDGEMENTS…………..……………………………………………………….….iii
ABSTRACT……………………….…………………………………………………………..…iv
TABLE OF CONTENTS….…………………..………………………………………………..…v
LIST OF TABLES….………………………………………………………………………..……x
LIST OF FIGURES…………………………………………………………………….………...xi
CHAPTER 1
1.0 INTRODUCTION…………………………………...…………………………….………… 1
1.3 Objectives…………..………….……………………………….………………….………5
1.5 Justification…………………..………………………………………..………………..... 6
2.0 Introduction………………………………………..……………………………………….9
2.5 Summary………………….……………………………………………………………..29
3.1.3 Trenching……………………...……………………………………..…………………..32
7
3.1.8 Steep Terrain…………………………….………………..…………………………….40
3.7.2 Algorithm…………………………………...………………………………………..54
8
3.7.4 Floyd-Warshall Algorithm………………………………………….…………………..57
3.7.5 Minimum-Connector………………………………………….…………………………57
3.9 Summary…………………………………………………………………………………61
4.0 Introduction…………………………………….……………………………………......62
4.2 Discussion…………………………...………………..……………………..…………77
4.3 Analysis……………………………………..………………..……………………….79
4.4.4 Ecology……………………………………..……..…………………………………83
4.4.5 Emissions………………………..……………….…………………………………..83
9
4.4.7 Socio-Economics……………..………………………………………………………84
5.0 Introduction………………………….………………………………………………86
5.1 Conclusion…………………….......………….………………………………………86
5.2 Recommendations……...…………………...………………………………………..87
REFERENCES……………………………………………………………………………….88
APPENDICES…………………...……………………………………………………………92
LIST OF TABLES
Table 4.2:Information1…………………………………………………………….…………....66
10
Table 4.9: Information 8…………………………………………………………………………73
LIST OF FIGURES
Figure 3.1 Pipeline Trenching Operations (Source: Photo courtesy of U.S. Pipeline)……….….34
Figure 3.2 Crew Lowering a Pipeline into a Trench (Source: Photo courtesy of U.S. Pipeline)..38
Figure 3.3 Horizontal Directional Drilling (HDD) Operation (Source: Photo courtesy of U.S.
Pipeline)……………………………………………..…………………………………..……….43
Figure 4.1 Pipe Network (in km) from Takoradi to all the regional capital towns in Ghana..…..63
11
Figure 4.3 Solution 2…………………………………………………………………………….68
12
CHAPTER 1
1.0 INTRODUCTION
Natural Gas is an essential source of energy all over the world. Ghana is one of the developing
countries in Africa whose dependants are increasingly depending on the use of Natural Gas. The
West Africa Gas Pipelines Project from Nigeria down to Ghana-Takoradi was developed to
facilitate the free flow of the Gas. The recent oil find in the west region of Ghana has contributed
to the production of the Natural Gas. This Gas has to be made available at a cheaper cost to all
consumers. This can only be done when the Gas is transmitted from the source to the consumers
via pipelines at a minimum cost. The pipelines are connected in network form to enable the free
flow of the natural gas from one place to another. There are several ways networks are used:
Telephone networks, Manufacturing and Distribution networks, and Computer networks. The
networking must be done efficiently to provide the consumers good service. To minimize these
The shortest path problems are the most fundamental and commonly encountered problems in
the study of transportation and communication network. All the networks are made up of nodes
(vertices) and arcs. One can move from one node along some edges to some other node. Since all
edges have length a shortest path is the one that has the smallest number of edges among all
paths. Similarly, a longest path is the one for which that sum is maximum. In some applications,
one requires not only the shortest path but also the second and the third shortest paths. The
research seeks to determine the shortest distances (and paths) from node to every other node in
the network.
13
Pipelines networks are not typically designed by transportation engineers. Perhaps, it is not
surprising then that the subject of pipelines has not received the required attention from
from inside the pipeline. Understanding flow in pipes is a common example of this focus. This
thesis cast a different light on the gas pipeline optimization problem by looking at it from the
perspective of an operation researcher and an industrial mathematician: How can the design of
Natural gas pipelines demand optimal design and layout since natural gas is one of the most
important energy sources used in the world. Normally, natural gas is supplied to users through
natural gas pipeline system .Many pipeline systems are built throughout the world , from the gas
reservoir to the end users in order to support the highly increasing demand for natural gas. In
America alone, from 1996 to 1998, at least 78 pipeline construction projects were completed
adding approximately 11.7 billion cubic feet per day of capacity (Energy Information
pipeline covering a total of 831.7 kilometers had been completed, and this is constantly being
Ghana is located in West Africa of which Togo, Burkina Faso and Cote d‟Ivoire form the
eastern, northern and western boundaries respectively. The south of it is the Gulf of Guinea.
Geographically, Ghana is divided into ten regions with their capital towns. The region and their
capital towns to which the gas pipelines are to be connected from the Western Region -Takoradi
are: Upper West-Wa, Upper East-Bolgatanga, Northern –Tamale, Brong -Ahafo -Sunyani,
The existing roads linking the regional capital towns will be used to provide access to the
constructing across paved roads, highways, railroads, steep terrain, water bodies and wet lands,
and when blasting through rock. All these are done so as to ensure safety and minimize cost and
distance covered.
The proposed connection of Gas Pipelines from Takoradi to other nine regional capitals is being
designed to make natural gas from the West African Gas Pipeline (WAGP) and the Jubilee Oil
Ghanaians are increasing depending on natural gas for its energy needs. In 2010 alone, the nation
consumed 124.1 billion cubic metres with demand of 165.2 billion cubic metres of natural gas,
forecast for 2015, Business Monitor International (BMI). Large portions of the gas consumed are
Ghana is yet to have a major natural gas production basins and the pipeline network of the gas is
not fully developed to other regional capitals. Most of the natural gas consumed are imported and
few supply from the West African Gas pipeline (WAGP). Of the natural gas consumed in Ghana,
0% is produced locally. There is no transmission of pipeline system connecting the West African
Gas Pipeline from Takoradi to other nine regional capitals. The primary function of the
transmission pipeline is to move huge amounts of natural gas hundreds of miles from the
producing region to the local natural gas utility delivery points. This delivery points, called “city
gates stations” are usually owned by distributions companies, although some are owned by
transmission companies. Compressor stations at required distances boost the pressure that is lost
through friction as the gas moves through the steel pipes (EPA 2000).
15
The Takoradi Thermal Power Station (TTPS) is located at Aboadze, 17 kilometres east of
Sekondi-Takoradi in the Western Region of Ghana. The Power Station which started operation in
1997 was initiated by the Volta River Authority to complement the existing Hydro Plant at
Akosombo and Kpong. TTPS is therefore a facility of strategic importance for meeting Ghana‟s
energy needs.
The current installed capacity of the Takoradi Power Plant is 550MW and is to be upgraded to
660MW by the CMS/VRA joint Venture. The wholly owned VRA portion of 330MW combined
cycle Plant with a 2x2x1 configuration meaning two (2) 110MW GE Frame 9E combustion gas
turbines, with two (2) HRSGs feeding one (1) 110MW Steam Turbine Generator. The voltage
output of the Generators which is 13.8kV is stepped up by Power Transformers to the national
With the objective of being an electrical supplier of choice for customers in Ghana and West
Africa, the Plant operates on two modes, namely Simple Cycle and Combined Cycle. The
primary fuel used for power generation is Light Crude Oil (LCO) which is normally received
from ocean tankers via a Single Point Mooring (SPM), connected to the plant by approximately
4.5km undersea pipeline and stored in four (4) 29,500m 3 capacity storage tanks. The secondary
fuel on site - Distillate Fuel Oil (DFO) is normally used for start-up and shutdown as it is less
volatile. On the average, a Gas Turbine at base load (110MW) can generate 2.5GW of electricity
The Plant has dual firing capacity and is being converted to run on natural gas. The West African
Gas pipelines company has also laid pipelines from the Niger Delta to the thermal plant to
16
The transmission segment of the gas industry is responsible for transporting natural gas from the
producer to the market areas via pipelines. The transmission segment is composed of pipelines,
compressor stations, city gate stations, and storage facilities. The energy sector of Ghana is faced
with urban and rural areas, transport sector, and industries total dependence of natural gas. This
gas is either not readily available or transportation is very high hence high cost of the gas. The
connection of gas pipelines from Takoradi to the other regional capitals seek to solve the
transportation problem. The installation of these pipelines across these regional capitals is
expensive and disruptive to the natural environment. Thus, this thesis seeks to find the shortest-
path these pipelines should be installed so as to reduce the cost of installation, minimize the
The demand for natural gas in Ghana has increased significantly as it has been a cheaper
alternative fuel used in our homes, industries, and mining and transportation sectors within
the country.
In view of this, the research is being conducted to connect the gas from Takoradi to the
regional capital towns to meet the growing demands of its citizenry. The study is therefore
conducted to design a minimum connection of gas pipelines that interconnects all the
17
1.3 OBJECTIVES
(ii) Extend gas pipelines from Takoradi to the regional capital towns in Ghana;
design.
Individuals, companies or organizations are required to monitor and adapt to the continuous
changes in the political, economic, social and technological environment. Consequently, it can be
said that a proper path and distances are crucial for the overall production and supply of natural
It is anticipated that the findings from the study will pave way for the Ghana National Petroleum
Corporation (GNPC) together with transportation engineers a clear picture of how to design a
network flow so as to minimize cost and distance covered. The study will also serve as a guide
for further research in other areas such as network design in transportation (rail lines, water
pipelines roads, etc), telecommunication (fibre-optic networks, computer networks etc) and rural
and urban electrification in a township or country. The result of the study will help the people
along the pipelines on environmental issues. Finally, the study serves as a partial fulfillment of
18
1.5 JUSTIFICATION
The rampant shortage of natural gas has been a problem to the consumers. The transport sector
has been affected most in this regard because of their increasing use of the gas. It takes several
days for the gas to be transported to the various regional capitals even if the gas is imported into
the country. This makes the system unfair to those who are far from the delivery point. This
situation calls for the connection of the gas pipelines from Takoradi to the other regional towns
This research is limited to all the ten regional capital towns in Ghana. The respondents will cover
some employees at the Gas industry, GPRTU, Ministry of Energy and Ministry of Roads and
Highways. The area covered was the map of road network from Takoradi to other regional
capital towns in Ghana. Findings of this study apply to the regional capital town which was
selected for the study. However, the findings could be adapted to other areas with similar
characteristics.
1.7 METHODOLOGY
Since the study is on minimum connection of gas pipelines from Takoradi to the regional capital
towns in Ghana, the researcher deployed the use of shortest path method. The shortest path
method include; Dijkstra‟s algorithm, Floyds-Warshall algorithm, Prism‟s algorithm etc. For the
purpose of this study, the researcher used Prism‟s algorithm to minimize the pipeline network
The research used a map provided by the Ghana Highway Authority. The map was the road
network system of the regional capital towns and the distance they cover within the region.
19
1.8 LIMITATION OF THE STUDY
The following are the limitation encountered when conducting the study;
(i) Financial constraints; as the researcher has to travel frequently to all the regional capital
towns.
(iii)Time constraint was a major setback in the conduct of the study since enough time was
The content of this thesis is arranged as follows. Chapter 1 establishes the introduction to
the network and pipeline optimization problem, background to the study, statement of the
methodology, limitation of the study, physical and network analysis and introduces
important gas pipeline basics . Chapter 2 presents relevant related research ideas which
have been published and shows their significance for the current research. Chapter 3
introduces, develops and illustrates the general algorithms of the method of solutions of
the problem. Chapter 4 analyzes optimal layout of the Gas Pipeline Project focusing on
Prim‟s algorithm for MST. Also in this chapter, the environmental damages revealed by
the pipeline network design were looked at. A summary, recommendations and
20
CHAPTER 2
LITERATURE REVIEW
2.0 INTRODUCTION
The review of literature summarizes a number of works in the field of networks, which have
developments in the pipeline industry. The study of networks, however, encompasses more than
just pipelines. Hence, the following sections contain a summary of published literature related to
transportation network optimization. They include developments in each of the analytical, exact
Accelerated growth in pipeline construction is expected due to recent industry deregulation in the
United States, West Africa etc. the growing presence of Mexico in the North America market is
analyzed in George and Mortenson (1995). They anticipate private investment in Mexico‟s
upstream sector followed by full integration with the continental natural gas market, in which
These developments indicate potential exists to improve the efficiency of gas pipelines systems
in North America at all levels from gathering systems through distribution. Wearmouth (1994)
found that, in 1993, 13 percent of Alberta‟s gas processing facilities or gas plants contained over
two thirds of the raw gas processing capacity in the providence. Additionally, over 50% of
Alberta‟s gas processing plants were designed to process less than ten percent (10%) of the total
gas processing requirements. These findings alone indicate room for improvement in gas
oil pipeline. Only physical elements of the pipeline designed were considered (pipeline diameter,
pumping stations, and suction and discharge pressures). The pipeline route was not included in
the optimizations parameters. Bolkan found little published research on optimal pipeline design
and determined that the common industry practice for design involves a trial-and-error approach
using cost benefit analysis for comparison of alternatives. Her findings are consistent with those
of the author in that little published research is available on pipeline optimization and little of
that relates to route design. There are few cases of published pipeline route optimization attempts
The North American Regional Gas Supply-Demand Model was commissioned by the Gas
Research Institute (GRI) in 1984 and is essentially a large-scale dynamic economic equilibrium
model (Nesbitt, 1988). The model incorporates all supply and demand regions in a network, the
links of which represent pipeline corridors. These corridors include a number separates pipelines
connecting similar regions. The simplified representation combined with the many difficulties
associated with the economy make the model subject to large errors. Its main value is in
government and regulatory agencies for guidance in broad-range policy making, but has limited
value for regional application. It is not a suitable model for determining the efficiency of the
The above publications touch the fringes of the problem of interest in this work that being the
spatially optimal design of pipeline networks. However, pipeline industries publications appear
to b e lacking I the specific area of interest. The previous chapter made an analogy between the
pipeline and electricity transportation systems. The latter of these systems were not of the focus
22
of the literature review as such may contain published research of some value to the pipeline
The factors influencing pipeline route selection are technical and engineering requirements,
environmental consideration and population density. However, these factors are chosen to
balance engineering and construction cost against environmental cost and future liability. The
engineering and technical considerations merge partly with some environmental considerations
used in this research include pipeline length, topography, surface geology, river and wetland
crossings, road and rail road crossings and the proximity to large proportion centers.
The literature shows that the route selection is the first and the most effective means of
preventing pipeline impacts from the outset. The literature is rich, diverse, detailed and helpful,
Environmental assessments of pipelines are neither new nor obscure nor precedent setting. The
conservations units and spill risks is standard in other projects by now. Of course, each pipeline
is different, but the methodologies and the types of impacts to be aware of have long since
Network optimization techniques have been studied extensively and applied in transportation,
manufacturing, communication and circuit board design industries to name a few. The problem
is usually this nature; how can a set of points be connected together with a system of links for the
least cost? In the case of pipeline networks, the points are supply and demand locations and the
links are pipelines. The similarities and differences between this problem and others which
23
utilize network optimization techniques are discussed. The overview highlights and development
transportation systems in rural areas of Nigeria. The authors used Odeda Local Government map
which has 88 villages connected by 96 roads. The project requires 388,270 meters to be covered.
Their findings shows that given the cost of fuel for travelling over 8000 meters to be 70.00 Naira
a transport company will make more profit with the implementation and use of Prim‟s algorithm;
taking into an account the cost of travelling over 8000 meters by an averagely good Toyota 14
sector bus. The solution obtained was reported and compared with the result obtained before they
apply Prim‟s algorithm on the existing network in both Table 2.2 and Table 2.3. For example, the
value of objectives function from Oguntoke to Olajugun was 544 Naira as shown in table 2.3
without using Prim‟s algorithm is applied. The same situation is to travel from Obantoko to
Nitoji with objective function of 507 Naira as against 186 Naira without using the algorithm.
They reported that the value of the objective function in Table 2.3 of column h is very low
compared to that of table 2.2 where Prim‟s algorithm is used. This, they affirm that the company
will make more profit with the implementation of the algorithm taking into account the cost of
travelling over 8000 meters is estimated. In the course of their work, they discovered that the
algorithm is very effective in providing shortest distances between two set of villages. In
addition, Arogundade and Akinwale found that in one way it reduces the cost of fuel and time for
transportation of passengers from one town to another which normally determines prices in
Nigerian transportation.
24
Table 2.1: Results of Objectives Function using Prim’s Algorithm
Dott (1997), in his study of Optimal Design of Natural Gas Pipeline of Amoco East Crisfield Gas
Pipeline project, (Alberta in Canada) containing 22 wells and connected by a gathering system of
25
approximately 66km pipeline uses Prim‟s and Steiner tree algorithm design a MST that covered
a 38km and less than 13.4% respectively of the original length of the pipeline. A similar study by
Bolkan (1991) in the same year of the Palliser Natural Gas Pipeline Project Canada which was
planned to connect 68 gas processing plant and compressor stations from Southeastern Alberta to
TransCanada pipeline and Foothills pipelines to consumers all over North America. The project
requires 709km laterals to a mainline approximately 243km in length for a total of 952km of
installed pipe and a cost of $365 million. The minimum spanning tree created by Prim‟s
algorithm for the Palliser Pipeline design length of 952km, for a 13% total length reduction. The
construction cost which corresponds to this 13% reduction may be approximated as 13% of $365
million or $46 million. The MST does not represent the shortest responsible network and up to a
further 13.4% reduction may be achieved with a Steiner tree. It is also important to note that the
Palliser Pipeline designers were probably unaware of or disinterested in any techniques which
could achieve this global or local minimum and thus the MST length is probably the more
appropriate measure.
Brimberg et al., (2003) studies the optimal design of an oil pipeline network for the South Gabon
oil field. This oil field has 33 nodes representing onshore platforms, onshore wells, several
connection points and one port (Gamba) with 129 possible arcs having total distance of 188.2
miles. Their finding in concert with Prim‟s algorithm reduces the connection to 156.2 miles
which is a reduction of 17% of the total distance to be covered. They reported that the
government saved a substantial amount of money from the number of pipelines used, labour
quantity, time and even minimize environmental destruction of the earth‟s topography.
The most recently published work in the field of network design promotes use of genetic
algorithm to find optimal or near-optimal solutions. Genetic algorithms use Darwin‟s theory of
26
evolution as a basis for their structure. Essentially, individuals seek out the fittest mate in order
to procreate and give their likewise fit offspring a greater chance of survival.
Each individual or solution has a unique genetic descriptor or chromosome. In this case, a
chromosome is a string of characters or genes each of which represents one element of the
“crossovers” or mating of the chromosomes. Sometimes mutation occurs which can reduce or
improve fitness of the offspring. Fitness of each solution is measured by a fitness function. Each
generation of solutions is usually of the same size as the previous generation with the less fit
solutions being replaced by more fit. The fit solution is always carried through until it is replaced
The genetic algorithm is superior to other heuristic algorithm because the crossing over and the
mutation chromosomes enable the offspring which can be very different from either parent.
Since a locally optimal tree network is likely to have a drastically different structure from the
The history of the Steiner tree problem presented here indicates that determining spatially
optimal networks is a complicated and imperfect science. In addition, the pipeline industry does
not prescribe too many of the solution principles, even simple ones for design. In real route-
design problems it is usually true that the finite number of routes is viable for consideration.
In computer science, Prim‟s algorithm is an algorithm that finds a minimum spanning tree for a
connected weighted undirected graph. This means it finds a subset of the edges that forms a tree
that includes every vertex, where the total weight of all the edges in the tree is minimized. Prim‟s
algorithm is an example of greedy algorithm. The algorithm was developed in the 1930 by Czech
27
mathematician Vojtẽch Jarnik and later independently by computer scientist Robert C Prim in
1957 and rediscovered by Edsger Dijkstra in 1959. Therefore it is also sometimes called the DJP
Gonnia et al., (2007) describe parallel implementation of Prim‟s algorithm for finding a
minimum spanning tree of a dense graph using MPI. Our algorithm uses a novel extension of
adding multiple vertices per iteration to achieve significant performance improvements on large
problems (up to 200,000 vertices). We describe several experimental results on large graphs
Gloor et al., (1993) described a system of visualizing correctness proofs of graph algorithms. The
system has been demonstrated for a greedy algorithm, Prim‟s algorithm for finding a minimum
spanning tree of undirected, weighted graph. We believe that our system is particularly
appropriate for greedy algorithms, though much of what we discuss can guide visualization of
proofs in other contexts. While an example is not a proof, our system provides concrete
examples to illustrate the operation of the algorithm. These examples can be referred to by the
user interactively and alternatively with the visualization of the proof where the general case is
portrayed abstractly.
McCarthy et al., (2009) presented the application of two well known graph algorithms,
Edmonds‟ algorithm and Prim‟s algorithm, to the problem of optimizing distributed SPARQL
queries. In the context of this paper, resolved by contacting any number of remote SPARQL
endpoints. Two optimization approaches are described. In the first approach, a static query plan
is computed in advance of query executions, using one of two standard graph algorithms for
finding minimum spanning trees (Edmonds‟ and Prim‟s algorithm). In the second approach, the
planning and the execution of the query are interleaved, so that as each potential solution is
expanded it is permitted to an independent query plan. Our optimization approach requires basic
28
statistic regarding RDF predicates which must be obtained prior to the user‟s query, through
Martel et al., (2002) studied the expected performance of Prim‟s minimum spanning tree (MST)
algorithm implemented using ordinary heaps. We show that this implementation runs in linear or
almost linear expected time on a wide range of graphs. This helps to explain why Prim‟s
algorithm often beats MST algorithms which have better worst-case run times. Specially, we
show that if we start with any n node edge graph and randomly permute its edge weights, then
Prim‟s algorithm runs in expected O (m + nlogn log (2m/n)) time. Note that O (m + nlogn log
(2m/n)) = O (m) when m = Ω (n log n log n). We extend this result to show that the same
expected run times apply even when an adversary can select the weights of m/logn edges and the
responsible weights of the remaining edges (which are then randomly assigned).
Chang et al., (2008) described the reason why it is beneficial to combine with graph theory and
board game. Forbye, it also descants three graph theories: Dijkstra‟s, Prim‟s, and Kruskal‟s
minimum spanning tree. Then it would describe the information about the game we choose and
how to combine the game with before-mentioned three graph theories. At all times, we could
account for the advantage of combining with these three graph theories and the game
specifically.
Dijkstra‟s algorithm conceived by Dutch computer scientist Edsger Djikstra in 1959 is a graph
search algorithm that solves the single-source shortest path problem for a path with nonnegative
edge path costs, producing a shortest path tree. This algorithm is often used in routing. An
For a given source vertex (node) in the graph, the algorithm finds the path with lowest cost (i.e.
the shortest path) between that vertex and any other vertex. It can also be used for finding cost
29
paths from a single to a single destination vertex by stopping the algorithm once the shortest path
to the destination vertex has been determined. For example, if the vertices of the graph represents
cities and edge path cost represent driving distance between pairs of cities connected together by
a direct road, Dijkstra‟s algorithm can be used to find the shortest route between one city and all
other cities. As a result, the shortest path first is widely used in network routing protocols, most
The literature makes it abundantly clear that the procedure commonly known today as Dijkstra‟s
Algorithm was discovered in the late 1950s, apparently independently, by a number of analysts.
There are strong indications that the algorithm was known in certain circles before the
publication of Dijkstra‟s famous paper. It is therefore somewhat surprising that this fact is not
Dijkstra, (1959) submitted his shorter paper for publication in Numerische Mathematik June
1959. In November 1959, Pollack and Wiebenson [1960] submitted a paper entitled Solutions of
the shortest route problem- a review to the journal Operations Research. This review briefly
discusses and compares seven methods for solving the shortest path problem.
However, the review presents a „highly efficient‟ method, attributed- as a dateless private
communication- to Minty. The procedure is defined as follows (Pollack and Wiebenson [1960, p.
225]: the objective is to find the shortest path from city A to city B.):
Sneidovich, (2006) described Dijkstra‟s Algorithm as one of the most popular algorithms in
computer science it is also popular in operations research. It is generally viewed and presented as
a greedy algorithm. In this paper we attempt to change this perception by providing a dynamic
algorithm is strongly inspired by Bellman‟s Principle of Optimality and that both conceptually
30
and technically it constitutes a dynamic programming successive approximation procedure par
excellence. One of the immediate implications of the perspective is that this popular algorithm
can be incorporated in the dynamic programming syllabus and in turn dynamic programming
Goldberg et al., (2006) proposed shortest path algorithms that use a search in combination with a
new graph-theoretic lower bounding technique based on landmarks and triangle inequality.
Our algorithms compute optimal shortest paths and work on any directed graph. We give
experimental results showing that the most efficient of our new algorithms outperforms previous
algorithms, in particular A search with Euclidean bounds, by a wide margin on road networks
Pangilinan et al., (2007) presented an overview of the multi-objective shortest path problem
(MSPP) and a view essential and resent issue regarding the methods to its solution. The paper
further explored a multi-objective evolutionary algorithm as applied to the MSPP and describes
solutions. Results show that the evolutionary algorithm can find diverse solutions to the MSPP in
polynomial time (based on several networks instances) and can be an alternative when other
As in the case of single –objective shortest path problem, the multi-objective shortest path
problem has been studied extensively by various researchers in the fields of optimizations, route
planning for traffic, transport design, information and communication networks design. The
MSPP is an extension of the traditional shortest path problem and it is concerned with finding a
set of efficient paths with two or more objectives that are usually in conflict. For example the
problem of finding optimal routes in communication networks involve minimizing delay while
31
maximizing throughput or finding efficient routes in transportation planning that
simultaneously minimize travel cost, path length and travel time. The concept of optimization in
optimization problem wherein the task is not to find a solution that optimizes a single objective
function. The task in a multi-objective problem is not to find an optimal solution for each
objective function but to find an optimal solution that simultaneously optimizes .all objectives.
And in most case, no single optimal solution exists, only a set of efficient and non- dominated
solutions.
A variety of algorithms and methods as dynamic programming, label selecting, label correcting,
interactive methods and approximation algorithms have been implemented and investigated with
respect to the MSPP. The problem is known to be NP-complete. It has been shown that a set of
problems exist wherein the number of Pareto-optimal solutions is exponential which implies that
any deterministic algorithm that attempts to solve it is also exponential in terms of runtime
complexity at least in the worst case. But some labels algorithm studies, disputes this exponential
behavior. They show that the number of efficient paths not exponential in practice. Other authors
avoid the complexity problem by developing new methods that run in polynomial time. For
instance, Hansen and Warburton separately develop Fully Polynomial Time Approximation
Schemes (FPTAS) for finding paths that are approximately Pareto-optimal. Interactive
procedures similarly avoid the problem of generating the complete set of efficient paths by
providing a user-interface that assists the decision maker to focus only on promising paths and
Given the wealth of literature in multi-objective algorithms for the MSPP, there still seems to be
a lack of reported review in Evolutionary Algorithm (EA) application in relation to the MSPP.
Several of the most recent alternative methods focus mostly on execution comparisons speed of
32
different MSPP algorithms but analysis of the salient issues in multi-objective performance
analysis such as runtime complexity, diversity, and optimally of non- dominated solutions are
almost omitted.
Divoky, (1990) presented a framework for solving the shortest-path, cost-flow problem with
procedure for a bigger problems. The efficiency of such shortest-path frameworks depend on the
technique employed to use the structure of the network from the solution of the bigger problem,
as well as on the efficiency of the frameworks. The topology of the networks is characterized by
arcs with positive weights and of sub-networks called zero-weight components. The techniques
for using topology include; identifying the zero-weight edges and adding them to the graph;
identifying and adding the basic sub-trees to the shortest-path tree; and interrupting the shortest-
path frameworks scanning or fixing the labels of all the nodes in the sub-tree.
Pettie et al., (2002) evaluated the practical efficiency of a new shortest path algorithm for
undirected graphs which was developed by the two authors. This algorithm works on the
Dijkstra‟s algorithm on sparse graphs for the all-pairs shortest path-problem, and more generally,
for the problem of computing single –source shortest path from different sources. Our extensive
experimental analysis demonstrated that this is also the case in practice. The authors presented
results which showed the new algorithm to run faster than Dijkstra‟s on a variety of sparse
graphs when the number of vertices ranges from a few thousand to a few million, and when
applications where the shortest paths and distances between only some specific pairs of origin-
destination (OD) nodes in a network are desired. The traditional repeated single-source shortest
path (SSSP) and all pair‟s shortest paths (APSP) algorithms often do unnecessary computation to
solve the MPSP problem. Our method is especially suitable for applications with fixed network
topology but changeable arc lengths and desired OD pairs. Preliminary computational
experiments demonstrate our algorithms superiority on airline network problems over other
Chang, (2009) described the shortest distance between two points as a straight line. But in the
real world, if those two points are located at opposite ends of the country, or even in different
neighborhood, it is unlikely you will find a route that enables you to travel from origin to
destination via one straight road. You might pull out a map to determine the fastest way to drive
somewhere, but these days, you are just as likely to use a Web-based service or a handheld
device to help with driving directions. The popularity of mapping applications for mainstream
consumer use once again has brought new challenges to the research problem known as
“shortest-path problem.”
The shortest path problem, one of the fundamental quandaries in computing and graph theory, is
intuitive to understand and simple to describe. In mapping terms, it is the problem of finding the
quickest way to get one location to another. Expressed more formally, in a graph in which
vertices are joined by edges and in which each edge has a value, or cost, it is the problem of
finding the lowest-cost path between two vertices. There are already several graph-search
algorithms that solve this basic challenge and its variations, so why is shortest-path perennially
34
Goldberg, (2001) principal researcher at Microsoft Research Silicon Valley, said there are many
“Shortest path in an optimization problem that‟s relevant to a wide range of applications, such as
network routing, gaming, circuit design, mapping,” Goldberg says. “The industry comes up with
new applications all the time, creating different parameters for the problem. Technology with
more speed and capacity, allows us to solve bigger problems, so the scope of the shortest-path
problem itself has become more ambitious. And now there are Web-based services, where
computing time must be minimized so that we can respond to queries in real time.”
algorithm. The blocked algorithm makes utilization of cache than does Floyd‟s original
algorithm. Experiments indicate that the blocked algorithm delivers a speedup (relative to the
unlocked Floyd‟s algorithm) between 1.6 and 1.9 on a Sun Ultra Enterprise 4000/5000 for graphs
that have between 240 and 1200 vertices is between 1.6 and 2.0.
Arulselvan et al., (2008) considered the problem of maximizing the total connectivity for a set o
wireless agents in a mobile ad hoc network. That is, given a set of wireless units having a start
point and a destination point, our goal is to determine a set of routes for the units which
maximize the overall connection time between them. Known as the COOPORATIVE
has several military applications including coordination of rescue groups, path planning for
unmanned air vehicles, geographical exploration and target recognition. The CCPM is NP-hard;
therefore heuristic development has been the major focus of research. In this work, we survey the
CCPM examining first some early combinatorial formulations and solution techniques. Then we
introduce new continuous formulations and compare the results of several case studies. By
35
removing the underlying graph structure, we are able to create a more realistic model of problem
Razzaque et al., (2009) said a fast algorithm is proposed to calculate kth power of an n × n
Boolean matrix that requires addition operations were p is the probability that an entry of the
matrix is 1. The algorithm generates a single set of inference rules at the beginning. It then select
entries (specified by the same inference rule) from any matrix (Ak-1) and adds them up for
modification of the proposed algorithm 1 can compute the diameter of any graph and for massive
required to find the first occurrence of 1 in a column in a linear search. The performance
comparisons say that the proposed algorithms outperform the existing ones.
The Floyd-Warshall algorithm, also variously known as Floyd‟s algorithm, the Roy-Floyd
algorithm, the Roy-Warshall algorithm, or the WFI algorithm, is an algorithm for efficiency and
simultaneously finding the shortest paths, (i.e., graph geodesics) between every pair of vertices
The Floyd algorithm is essentially equivalent to the transitive closure algorithm independently
discovered by Roy (1959) and Warshall (1962) (Pemmaraju and Skiena 2003), which is the
Hougardy, (2010) the Floyd Warshall algorithm is a simple and widely used algorithm to
compute shortest paths between all pairs of vertices in an edge weighted directed graph. It can
also be used to detect the presence of negative cycles. Hougardy, (2010) show that for this task,
many existing for the implementation of the Floyd- Warshall algorithm will fail because
Chen et al., (2007) focused on the optimization problems about complicated network, this paper
present an algorithm KSPA to solve the K-shortest paths problem complicated networks based
36
on algorithm, in which the time cost is taken as target function and the establishment of the target
function model is given. Experimental results show that the proposed KSPA maintains an
excellent efficiency on certain public traffic data. It can be used to solve the K-shortest paths
problems in multi-graph.
Lysgaard, (2000) presented a new algorithm for finding the shortest path from a source to a
single sink in a network, in which the location in the plane of each node is known. The algorithm
consists of two phases. In the first phase a heuristic solution is to the shortest path problem is
found. In the second phase the upper bound provided by the heuristic solution is utilized in a
modification of a standard shortest path algorithm. Estimates based on computational test show
that on average the computation time of the presented algorithm is on the order of 40-60% of the
Rabbani et al., (2008) presented a distribution network design problem in a multi-product supply
chain system that involves the locating production plants and distribution warehouse as well as
determining the best strategy for distributing the product from plants to warehouse and from
warehouse to customers. The goal is to select the optimum numbers, locations and capacities of
plant and warehouses to open, so that all customers demand of all product types are satisfied at
minimum total costs of the distributing network. Unlike most of the previous researches, our
mathematical programming model for designing a supply chain distribution network. Finally,
this paper represents a real-case study to investigate designing a pharmaceutical supply chain
Multi pairs shortest path problem (MPSP) that arises in many applications where the shortest
paths and distances between only some specific pairs of origin destination (OD) nodes in a
network are desired. The traditional repeated single-source shortest path (SSSP) and all pairs
37
shortest path (APSP) algorithm often do unnecessary computation to solve the MPSP problem.
Sololá et al., (2005) proposed a new shortest path algorithm to save computational work when
solving the MPSP problem. Our method is especially suitable for applications with fixed network
topology but changeable arc length and desired OD pairs. Preliminary computational
experiments demonstrate our algorithm superiority on airline network problems over other APSP
Hseih et al., (2004) designed shortest path routing algorithms is in general more difficult than
designing simple routing algorithms. In this paper we derived a shortest path routing algorithm
for pyramid networks. The proposed algorithm takes O (1) time to determine the shortest path
between any two nodes in a pyramid network. We also designed a distributed routing algorithm
such that an intermediate node takes O (1) time to confirm the next node along the shortest path
Algorithms for finding the shortest path from one point to another have been researched for
years. Applications abound, but let‟s keeping things simple by saying we want to find the
shortest path from one point A to point B in a city with just a few streets and interactions. There
are quite a few algorithms that have been developed to solve such problems, all with different C;
One of the fastest algorithms for solving this problem has a runtime of
where E is the number of segments, and V is the number of intersections. To put this in
perspective, the algorithm will take about 2 seconds to find the shortest path in a city with 10000
intersections and 20,000 road segments (there are usually about 2 road segments per
intersection). The algorithm, known as Dijkstra‟s Algorithm, is fairly complex, and requires the
use of the data structure known as a priority queue. In some applications, however, even this
38
runtime is too slow (consider finding the shortest path from New York City to San Francisco-
there are millions of intersections in the US), and programmer try to do better by using what are
and is often computed by an algorithm of its own. In the shortest path problem, for example, it is
useful to know approximately how far a point is from the destination. Knowing this allows for
the development of faster algorithm (such as A*, an algorithm that can sometimes run
significantly faster than Dijkstra‟s algorithm) and so programmers come up with heuristics to
approximate this value. Doing so does not always improve the runtime of the algorithm in the
worst case, but it does make the algorithm faster in most real world applications.
Traffic information systems are among the most prominent real-world applications of Dijkstra‟s
algorithm for shortest paths. Schulz et al, (2000) considered the scenario of a central information
server in the realm of public railroad transport on wide area networks. Such a system has to
process a large number of on-line queries for optimal travel connections in real time. In practice,
this problem is usually solved by heuristic variations of Dijkstra‟s algorithm, which do not
guarantee an optimal result. We report results from a pilot study, in which we focused on the
travel time as the only optimization criterion. In this study, various speed-up techniques for
Dijkstra‟s algorithm were analyzed empirically. This analysis was based on the timetable data of
In graph theory, the shortest path problem is the problem of finding a path between two vertices
(nodes) such that the sum of the weights of its constituent edges is minimized. An example is
finding the quickest way to get from one location to another on a road map; in this case, the
vertices represents the location and the edges represents the segments of road and are weighted
39
Formally, given a weighted graph (that is, a set V of vertices, a set E of edges, and a real-valued
The problem is also sometimes called the single-pair shortest path problem, to distinguish it from
The single-source shortest path problem, in which we have to find shortest paths from a
The single-destination shortest path problem, in which we have to find shortest paths
from all vertices in the graph to a single destination vertex v. this can be reduced to the
The all-pairs shortest path problem, in which we find shortest paths between every pair of
These generalizations have significantly more efficient algorithms than the simplistic approach
40
2.4 SUMMARY
industry. They include developments in each of the analytical, computational and heuristic
solution approaches. In the next chapter, we shall introduce, develop and illustrate the general
41
CHAPTER 3
RESEARCH METHODOLOGY
3.0 INTRODUCTION
This chapter focuses on the method and algorithms for solving the problem of designing gas
pipeline network. The techniques employ here are useful for getting several alternative systems
of near-optimal arrangements which can be compared using economic and qualitative analysis
Network design methods are not well recorded and applied in practice. The practical example
consider in this chapter will provide guidelines for effective pipeline network design for
industries that uses network in its operations. Include is the environmental factors in the network
Shortest path problems are the most fundamental and most commonly encountered problem in
the study of transportation and communication networks. The shortest path problem arises when
trying to determine the shortest, cheapest or the most reliable path between one or many pairs of
optimization problems such as vehicle routing and network design often called the solution of
large number of shortest path problems as subroutines. Consequently, designing and testing
efficient algorithm for the shortest path problem has been a major area of research in network
optimization.
42
3.1 GENERAL PIPELINE CONSTRUCTION PROCEDURES
Before construction starts, engineering surveys are conducted of the ROW centerline and extra
workspaces, and complete land or easement acquisition of private and state lands is finalized. If
the necessary land rights or easements are not obtained through good faith negotiations with
landowners and the project is approved by the FERC, the pipeline company could use the right
of eminent domain granted it an easement under Section 7(h) of the Natural Gas Act (NGA). The
pipeline company is still required to compensate the landowners for the Right(s) –of –Way
(ROW), as well as for any damages incurred during construction. The pipeline company
generally pays the market price for the property. However, the level of compensation is
determined by the court system according to state laws regarding eminent domain. Eminent
domain is used only as a last resort, because the process can take up to 2 to 3 years (FERC 2007).
The landowner normally is compensated a fair market value for a permanent easement, which
typically allows the landowner continued use and enjoyment of the aboveground property, with
some limitations. The limitations typically prohibit excavation as well as the placement of
structures and trees within the easement to preserve safe access for maintenance equipment when
necessary and allow for uninhibited aerial inspection of the pipeline system.
The landowner is generally compensated at an amount lower than fair market value when the
pipeline company needs only a temporary construction easement, since this land reverts back to
the landowner after construction for full use and enjoyment without any restrictions, although
Additionally, landowners are compensated for any damages or losses they may incur, such as the
loss of crop revenues, as a result of construction across their property. Normally, they are
43
Overland pipeline construction in a rural environment generally would proceed as a moving
assembly line, as summarized below. Typically, job-specific work crews would construct the
Standard pipeline construction is composed of specific activities, including survey and staking of
the ROW, clearing and grading, trenching, pipe stringing, bending, welding, lowering-in,
pipeline construction methods, the pipeline company would use special construction techniques
where warranted by site-specific conditions. These special techniques would be used when
constructing across rugged terrain, water bodies, wetlands, paved roads, highways, and railroads.
The first step of construction involves marking the limits of the approved work area (i.e.,
construction ROW boundaries, additional temporary workspace areas) and flagging the locations
of approved access roads and foreign utility lines. Wetland boundaries and other environmentally
sensitive areas also are marked or fenced for protection at this time. Before the pipeline trench is
Before clearing and grading activities are conducted, the landowners‟ fences (if any) are braced
and cut, and temporary gates and fences are installed to contain livestock, if present. A clearing
crew follows the fence crew and clears the work area of vegetation and obstacles
(e.g., trees, logs, brush, rocks). Grading is conducted where necessary to provide a reasonably
level work surface. Root stock is left in the ground in areas where the ground is relatively flat
and does not require grading. More extensive grading is required in steep side slopes, vertical
The trench is excavated to a depth that provides sufficient cover over the pipeline after
backfilling. Typically, the trench is about 4 to 6 feet wide in stable soils and about 2 to 5 feet
deep to the top of the pipe, depending on the pipeline‟s diameter. This depth allows for the
required minimum of 30 to 36 inches of cover. Additional cover for the pipeline is provided at
road and water body crossings, while less cover (a minimum of 18 inches) is required in rock.
The trenching crew uses a wheel trencher or backhoe to dig the pipe trench.
When rock or rocky formations are encountered, tractor-mounted mechanical rippers or rock
trenchers are used to fracture the rock prior to excavation. In areas where mechanical equipment
could not break up or loosen the bedrock, blasting is required. The contractor would be required
to use explosives in accordance with state and federal guidelines to ensure a safe and controlled
blast. Excavated rock would then be used to backfill the trench to the top of the existing bedrock
profile.
In areas where there is a need to separate topsoil from the subsoil, the topsoil is graded prior to
trenching. The topsoil over the ditch line is segregated for the majority of the project (unless
requested otherwise by the landowner). Clearing activity on the spoil side is limited to what is
necessary for construction activity. Topsoil is stored in a pile that is separate from the subsoil to
allow for proper restoration of the soil during the backfilling process. Spoil typically is deposited
on the nonworking side of the ROW, and gaps are left between the spoil piles to prevent storm
water runoff from backing up or flooding. Topsoil is returned to its original ground level plus
some mounding to account for soil subsidence after the subsoil is backfilled in the trench. As
backfilling operations begin, the soil is returned to the trench in reverse order, with the subsoil
put back first, followed by the topsoil. This process ensures that the topsoil is returned to its
original position.
45
Figure 3.1 Pipeline Trenching Operations (Source: Photo courtesy of U.S. Pipeline)
Prior to or following trenching, sections of externally coated pipe up to 80 feet long (also
referred to as “joints”) are transported to the ROW by truck over public road networks and along
authorized private access roads and placed, or “strung,” along the trench in a continuous line.
After the pipe sections are strung along the trench and before joints are welded together,
individual sections of the pipe would be bent where necessary to allow for fitting the pipeline
uniformly with the varying contours of the bottom of the trench. Workers would use a track-
mounted, hydraulic pipe-bending machine to shape the pipe to the contours of the terrain. The
46
bending machine uses a series of clamps and hydraulic pressure to make a very smooth,
controlled bend in the pipe. All bending must be performed in strict accordance with federally
prescribed standards to ensure the integrity of the bend. When a section of pipe requires multiple
or complex bends, that work is performed at the factory, or else pipeline fittings such as elbows
are installed.
Welding is the process that joins the various sections of pipe together into one continuous length.
After the pipe sections are bent, the joints are welded together into long strings and placed on
temporary supports. The pipe gang and a welding crew are responsible for the welding process.
The pipe gang uses special pipeline equipment called side booms to pick up each join of pipe,
align it with the previous joint, and make the first part (a pass called the stringer bead) of the
weld. Additional filler passes are made by welders who immediately follow the stringer bead on
what is called the welding firing line. Stringer, hot-pass, and capping welders make up the firing
line, and they are followed in certain locations by tie-in welders. (On difficult-fit welds, the
welder sometimes also back-welds the pipe by welding the welds from the inside to assure the
integrity of the weld) The pipe gang then moves down the line to the next section and repeats the
process. The welding crew follows the pipe stringing gang to complete each weld.
In recent years, contractors have used semiautomatic welding units to move down a pipeline and
complete the welding process. Semiautomatic welding must be completed to strict specifications
and still requires qualified welders, and personnel are required to set up the equipment and hand-
As part of the quality assurance process, each welder must pass qualification tests to work on a
particular pipeline job, and each weld procedure must be approved for use on that job in
accordance with federally adopted welding standards. Welder qualification takes place before the
project begins. Each welder must complete several welds using the same type of pipe as that to
be used in the project. The welds are then evaluated by placing the welded material in a machine
47
and measuring the force required to pull the weld apart. Interestingly, the weld has a greater
tensile strength than the pipe itself. The pipe must break before the weld.
One hundred percent of the welds undergo radiographic inspection (X-ray). A second quality
assurance test ensures the quality of the ongoing welding operation on-site. In this test, qualified
technicians take X-rays of the pipe welds to ensure that the completed welds meet federally
prescribed quality standards. The X-ray technician processes the film in a small, portable
darkroom at the site. If the technician detects any flaws, the weld is repaired or cut out, and a
new weld is made. Another type of weld quality inspection employs ultrasonic technology.
A protective epoxy coating or mastic is applied to the welded joints once the welds are approved.
Line pipe receives an external coating, which inhibits corrosion by preventing moisture from
coming into direct contact with the steel. This process is normally completed at the coating mill
where the pipe is manufactured or at another coating plant location before it is delivered to the
construction site. All coated pipes, however, have uncoated areas 3 to 6 inches from each end of
the pipe to prevent the coating from interfering with the welding process. Once the welds are
made, a coating crew coats the field joint, the area around the weld, before the pipeline is
Pipeline companies use several different types of coatings for field joints, the most common
being fusion-bond epoxy, polyethylene heat-shrink sleeves, or heated mastic tape. Prior to
application, the coating crew thoroughly cleans the bare pipe with a power wire brush or a
sandblast machine to remove any dirt, mill scale, or debris. The crew then applies the coating
and allows it to dry prior to lowering the pipe in the ditch. Before the pipe is lowered into the
trench, the coating of the entire pipeline is inspected to ensure it is free of defects. The pipeline is
then electronically inspected, or “jeeped,” for faults or voids in the epoxy coating and visually
inspected for faults, scratches, or other coating defects. Damage to the coating is repaired before
Before the pipeline is lowered into the trench, an environmental inspector inspects the trench to
be sure it is free of livestock or wildlife that may have become trapped in the trench, as well as
free of rocks and other debris that could damage the pipe or protective coating. At the end of the
day after welding is completed, the pipe crew installs end caps (rubber expandable plugs) at the
end of the pipeline to prevent debris and wildlife from entering the pipe. In areas where the
trench had accumulated water since being dug, dewatering could be necessary to allow
inspection of the bottom of the trench. The pipeline then is lowered into the trench. On sloped
terrain, trench breakers (stacked sandbags or foam) are installed in the trench at specified
In rocky areas, the pipeline is protected with a rock shield (a fabric or screen that is wrapped
around the pipe to protect it and its coating from damage by rocks, stones, roots, and other
debris) or sand aggregate. In an alternative method, the trench bottom is filled with padding
material (e.g., finer grain sand, soil, or gravel) to protect the pipeline. Topsoil is not used as
padding material.
Lowering the welded pipe into the trench demands the close coordination of skilled operators.
By using a series of side-booms (tracked construction equipment with a boom on the side),
operators simultaneously lift the pipe and carefully lower the welded sections into the trench.
Nonmetallic slings protect the pipe and its coating as it is lifted and moved into position.
The trench is then backfilled using the excavated material. As with previous construction crews,
the backfilling crew takes care to protect the pipe and coating as the soil is returned to the trench.
The soil is returned to the trench in reverse order, with the subsoil put back first, followed by the
topsoil. The segregated topsoil is restored to its original grade and contour last by using either a
remove rocks, bring in clean fill to cover the pipe, or cover the pipe with a material to protect it
from sharp rocks. Once the pipe is sufficiently covered, the coarser soil and rock can be used to
Figure 3.2 Crew Lowering a Pipeline into a Trench (Source: Photo courtesy of U.S.
Pipeline)
In addition to standard pipeline construction methods, special construction techniques are used
when warranted by site-specific conditions, such as when constructing across paved roads,
highways, railroads, steep terrain, water bodies, and wetlands, and when blasting through rock.
Additional construction areas, or temporary extra workspaces, are required for construction at
road crossings, railroad crossings, crossings of existing pipelines and utilities, stringing truck
turnaround areas, wetland crossings, horizontal directional drilling (HDD) entrance and exit pits,
and open-cut water body crossings. These extra workspaces are located adjacent to the
50
construction ROW and could be used for such purposes as spoil storage, staging, equipment
movement, material stockpiles, and pull-string assembly associated with HDD installation.
Individual extra workspaces would range in size from less than 0.1 to 2 acres and would be
returned to their preconstruction condition and former use following completion of construction
activities.
Construction across paved roads, highways, and railroads would be carried out in accordance
with the requirements of the appropriate road and railroad crossing permits and approvals
obtained by the pipeline company. In general, major paved roads, highways, and railroads are
crossed by boring beneath the road or railroad. Boring requires excavating a pit on each side of
the feature, placing the boring equipment in the pit, and then boring a hole under the road at least
equal to the diameter of the pipe. Once the hole is bored, a prefabricated pipe section is pushed
through the borehole that would consist of either extra-heavy wall-thickness carrier pipe or two
pipes consisting of an outer casing pipe and the inner carrier pipe. For long crossings, sections
could be welded onto the pipe string just before being pushed through the borehole. Boring
crossings. Each boring project is expected to take 2 to 10 days. Operations typically are
conducted 24 hours per day, 7 days per week until the boring is completed.
Smaller unpaved roads and driveways are crossed using the open-cut method where permitted by
local authorities or private owners. The open-cut method requires temporarily closing the road to
traffic and establishing detours. In instances where a reasonable detour is not feasible, at least
one lane of traffic is kept open except during brief periods when it is essential to close the road to
install the pipeline. Most open-cut road-crossing construction projects (including road
resurfacing) take some weeks to complete, depending on soil settlement after compaction. (In
general, most pipeline companies prefer to wait several weeks before final resurfacing.) Posting
51
signs at open-cut road crossings and other measures are undertaken to help ensure safety and
Additional grading may be required in areas where the proposed pipeline route crosses steep
slopes. Steep slopes often need to be graded down to a gentler slope to accommodate pipe-
bending limitations. In such areas, the slopes are cut away and, after the pipeline are installed,
In areas where the proposed pipeline route crosses laterally along the side of a slope, cut-and-fill
grading may be required to obtain a safe, flat, work terrace. Generally, on steep side slopes, soil
from the high side of the ROW is excavated and moved to the low side of the ROW to create a
safe and level work terrace. Under these circumstances, the topsoil is stripped from the entire
width of the ROW. After the pipeline is installed, the soil from the low side of the
ROW is returned to the high side, the topsoil is replaced, and the slope‟s original contours are
restored.
In steep terrain, temporary sediment barriers such as silt fences and certified weed-free straw
bales are installed during clearing to prevent the movement of disturbed soil off the ROW.
Temporary slope breakers that consist of mounded and compacted soil are installed across the
ROW during grading, and permanent slope breakers are installed during cleanup. Following
construction, seed is applied to steep slopes, and the ROW is mulched with certified weed-free
hay or non brittle straw or is covered with erosion-control fabric. Sediment barriers are
The generally preferred method of crossing a water body that is flowing at the time of
construction is HDD compared to the open-cut method, since HDD‟s decreasing cost and lack of
environmental impact are making it more popular. The open-cut crossing method involves
52
trenching through the water body while water continues to flow through the trenching area. If no
water is flowing at the time of construction, the water body is crossed using conventional upland
The open-cut crossing method involves excavating a trench across the bottom of the river or
stream to be crossed with the pipeline. Depending on the depth of the water, the construction
equipment may have to be placed on barges or other floating platforms to complete excavation of
the pipe trench. If the water is shallow enough, the contractor can divert the water flow with
dams and flume pipe, which can allow backhoes working from the banks or the streambed to dig
the trench.
The contractor prepares the pipe for the crossing by stringing it out on one side of the stream or
river and then welding, coating, and hydrostatically testing the entire pipe segment. Side booms
carry the pipe segment into the stream bed just as they do for construction on land, or the
construction crew floats the pipe into the river with flotation devices and positions it for being
buried in the trench. Concrete weights or concrete coating ensure that the pipe will stay in
position at the bottom of the trench once the contractor removes the flotation devices.
The flume, dam-and-pump, and HDD methods also could be considered as alternative crossing
methods. The flume crossing method involves diverting the flow of water across the trenching
area through one or more flume pipes placed in the water body. The dam-and-pump method is
similar to the flume method except that pumps and hoses are used instead of flumes to move
The HDD method involves drilling a hole under the water body and installing a prefabricated
segment of pipe through the hole. Before a directional drill is designed, core samples are taken
on both sides of the crossing to evaluate the underground rock and sand formations. If the
subsurface will support a directional drill, the engineer can design a crossing that establishes the
53
entry and exit points of the pipeline crossing and its profile as it would traverse under the
crossing.
The HDD method involves drilling a pilot hole under the water body and banks and then
enlarging the hole through successive reaming until the hole is large enough to accommodate a
prefabricated segment of pipe. Throughout the process of drilling and enlarging the hole, slurry
made of nontoxic fluids (e.g., bentonite and water) is circulated through the drilling tools to
lubricate the drill bit, remove drill cuttings, and keep the hole open. This slurry is referred to as
drilling mud.
While this drilling is in progress, the line pipe sections are strung out on the far side of the
crossing for welding. Once welded, the joints are X-rayed, coated, hydrostatically tested, and
then placed on rollers or padded skids in preparation for being pulled through the drilled-out
hole.
Figure 3.3 Horizontal Directional Drilling (HDD) Operation (Source: Photo courtesy of
U.S. Pipeline)
54
3.3 ABOVEGROUND FACILITY CONSTRUCTION PROCEDURES
The aboveground facilities are constructed concurrent with pipeline installation, but construction
is conducted by special fabrication crews that generally work separately from the pipeline
construction spreads.
Construction of the compressor stations involves clearing, grading, and revising the sites to the
surveyed elevations where necessary for placement of concrete foundations for buildings and to
support skid-mounted equipment. Prefabricated segments of pipe, valves, fittings, and flanges
are welded at the shop or on-site and assembled at the compressor station site. The compressor
units and other large equipment are mounted on their respective foundations, and the equipment
is micro-leveled to reduce vibration; then the compressor enclosures are erected around them.
exhaust stack silencers, and air inlet silencers) and emission control technology are installed as
needed to meet applicable federal, state, and/or local standards. Electrical, domestic water and
Facility piping, both above and below ground, are installed and hydrostatically tested before
being placed into service. Controls and safety devices such as the emergency shutdown system,
relief valves, gas and fire detection facilities, and other protection and safety devices are also
checked and tested. Upon completion of construction, all disturbed areas associated with the
aboveground facilities are finish-graded and seeded or covered with gravel, as appropriate. All
roads and parking areas are graveled. Additionally, the compressor station sites are fenced for
Somewhat less than 100 workers and five inspectors are required to construct a typical
compressor station. Initial site preparation typically takes approximately 16 to 20 weeks, while
55
actual installation requires more than 6 months. The lead time needed to purchase some
Direct emissions would result from the construction of a natural gas compressor station; although
construction impacts are expected to be temporary and transient, and the short-term exposure
levels are considered minimal. These emissions include exhaust from the construction equipment
and vehicle engines and fugitive dust from the disturbed areas along the ROW.
Carbon monoxide emissions are emitted in the largest quantities during construction, followed
Physical networks are perhaps the most common and the most readily identifiable class of
networks. Indeed, often when we think of networks, we typically envision a physical network, be
chip. Table 3.1 summarizes the components of any physical networks that emerge in this varied
Nodes Arcs
satellites
56
reservoirs, lakes hydraulic fluids
circuits processors
Natural gas network analysis is used to determine the amount of gas, flow rate of the gas and the
pressure used in the pipeline system. Besides that, it can also be used to determine the usage of
gas by consumers especially during the peak times. By making the network analysis, time of
maintenance and failure of pipeline can be measured. The effect of new load on the pipeline can
also be predicted. Generally, network analysis can be divided into three types; via Network
Tracing, Network Routing and Network Allocation. Network tracing is done in other to trace the
allocation of the particular pipeline in the system. Network routine is done in order to find the
optimum path or the shortest path which cost less for the system. Network allocation is done in
order to locate all the pipelines into the main support. Our analysis of the network shall be
Route networks are composite entities that frequently model complex distribution and logistics
with an inventory of goods at its warehouse must ship to geographically dispersed retail centers,
each with a given customer demand. Each arc connecting a supply point to a retail centre incurs
57
cost based upon some physical network, in this case the transportation network. Rather than
solving the problem directly on the physical network, we preprocess the data and construct
transportation routes. Hence, in effect an arc connecting a supply point and a retail centre might
(ii) From the rail station to a rail head elsewhere in the system,
(iv) From the distribution centre (on a local delivery truck) to the final consumer.
If we assign the arc the composite distribution cost of all the intermediary legs, as well as the
distribution capacity for this route, this problem becomes a classic network transportation model
defined over a bipartite graph: find the flow from the gas reserves to consumers that minimizes
overall costs.
Similar application arises in all of the problem settings we have mentioned in our discussion of
physical networks. In some applications, however the network might contain intermediary points
that serve as transshipment nodes; they neither generate flow nor consume flow. Pumping
stations in a natural gas transportation network would be an example. In this application, the
network model is a general minimum cost flow problem, rather than a classic transportation
Metering stations are placed periodically along interstate natural gas pipelines. These stations
allow pipeline and local distribution companies to monitor, manage, and account for the
natural gas in their pipes. Essentially, these metering stations measure the flow of gas along the
pipeline, allowing pipeline companies to track natural gas as it flows along the pipeline.
58
Metering stations employ specialized meters to measure the natural gas as it flows through the
pipeline without impeding its movement. In essence, the metering station is the company‟s “cash
Meter/regulator stations are generally constructed adjacent to the cleared pipeline right-of-way
(ROW) at each of the receipt and interconnect points to meter the flow and adjust the pressure of
natural gas received from or delivered to those systems. A meter/regulator station typically
includes meter and regulator equipment, a filter separator, odorant equipment, and a control
Natural gas is produced from wells drilled into hydrocarbon-bearing geological structures
beneath the surface of the earth. Natural gas I s composed largely of methane and usually
contains a limited amount of impurities such as oxygen, nitrogen, carbon dioxide and sometimes
hydrogen sulfide which are natural by-products of gas production (Encarta encyclopedia, 2007)
59
Natural gas is transported from the well, where it is produced to the burner-tip of the end user
through a number of interconnected pipeline networks. Because of its gaseous nature, natural gas
must be transported through one completely interconnected pipeline system to avoid leakage to
the atmosphere. The system which transports this gas is comprised of a number of networks
Natural gas produced from the wells is collected and shipped into a processing plant where
moisture and impurities are removed from the gas before it is shipped farther downstream. The
network connecting a number of gas wells to a processing plant is called a gathering system. In
this arrangement the plant acts as a hub. A hub is a central facility within a network. The plant or
hub receives incoming gas from gathering pipes and feeds a smaller number of processed
streams existing in the plant. Some of these processed streams might include processed gas,
Processed gas from many plants is collected by a larger pipeline system, sometimes called a
trunk line. A trunk line must have a number of destinations to which processed or sales gas is
delivered. At this time in Escravos-Lagos, the Escravos-Lagos Pipeline (ELP) is the main gas
trunk line and collects most of the sales gas within the state.
Distribution of the natural gas to most residential, commercial and industrial consumers is done
systems are similar to gather systems except they usually much larger in scale and homes and
businesses are connected to the network instead of the gas wells or fields.
60
3.5.3 GENERAL UNIQUE ASPECTS OF GAS PIPELINE NETWORKS
Pipeline transportation is unique in the transportation industry. The following four sub-sections
describe some of the important differences which make pipeline networks worthy of special
Natural gas is captive to the pipeline mode of transportation because of its gaseous state.
Liquefied Natural Gas (LNG) can be transported via truck or ship, but LNG is expensive to
produce and must be shipped at a very low temperature making this option prohibitively in most
cases.
There are some difficulties in the use of natural gas. The first is the availability: 73% of the
world‟s known reserves are in West Asia (Bangladesh), further east (Myanmar), West Africa
(Nigeria) and former Soviet Union states (BP 1998). In addition to this, its low energy density
(especially vis-à-vis oil) makes it difficult to store and transport. While oil can easily be shipped
by tankers, gas must typically be transported through pipelines, which are capital- intensive.
Pipeline technology is well established relatively mature. For long- distance transmission, natural
gas is transported through high-pressure (900+p.s.i) pipelines with diameters typically in the
three to four (3-4) foot ranges (Tongia, Arunnachalam. et.al, 1998). A transmission system
consists of not only pipe to transport the gas but also of compressor stations- to re-pressurize the
gas which loses pressure due to friction- at regular intervals. A small fraction of the transported
gas is consumed by the gas-driven compressors, though it is possible to use electric compressors
instead. The most notable improvements in technology have been for superior materials and
advanced welding techniques that have made pipelines thinner walled, safer, and less expensive
than before.
61
The other method of transport, as Liquefied Natural Gas (LNG), is even more capital intensive.
This is because natural gas liquefies only at very low temperatures, below- 160 degree Celsius.
For a six million (6,000,000) tons/year system (about 8 BCM/year re-gasified), liquefaction
plants, LNG tankers, receiving and storage terminals and re-gasification facilities can typically
cost over four billion dollars. Because of this, for relatively for short distances- under about 2000
miles-pipelines are less expensive than LNG (Energy Information Administration, 1997b).
Strong economies of scales exist for natural gas pipelines, more so than for LNG facilities. The
volumes economically transported by a pipeline strongly depend on the pipeline diameter. Due to
the non linear effects of friction and pressure losses, doubling the diameter of the pipe allows for
about six times the gas flow. Typical transmission pipelines operate most economically at
Et.al.1998). Mode captivity in the gas transportation industry impacts the gas pricing directly. As
with transportation of goods in general, improved efficiency in gas shipping this industry may
not incorporate efficiency improvements as quickly as might occur in more competitive areas.
Pipeline networks are continuous structures which can receive or deliver gas volumes at any
point along the route so long as the necessary mechanical components exist at the location. These
components include valve and a metering device with downstream pipeline connections.
Continuous transportation systems contrast with discrete systems in which receipts and deliveries
may only be made at certain drop-off and pick-up locations. An example of this is an airline
route in which passengers and baggage may only board or exist an air plane while it is landed at
an airport.
62
3.5.6 TIME OF GAS TRANSPORT
In the gas pipeline industry, time of travel is of no consequence. Because gas receipts are
blended along the length of the pipeline, pipeline operators allow shippers to benefit from
The unimportance o gas transport time does not; however imply that there is not some cost
associated with transporting gas longer distance. It is true that consumers who live farther from
the geological supply of natural gas pay higher prices for their gas than consumers close to gas
production areas. This additional cost however is borne out of pipeline infrastructure and its
operation, but not from actual time of transport from well-head to burner-tip.
Pipelines networks are largely directed networks. A directed network is composed of links which
flow in only one direction. An undirected network encompasses flow in both directions on its
links. An urban road system which utilizes no one street is an example of an undirected network.
Pipelines systems tend to be directed because they connect supply and demand centres, and flow
is generally from supply to demand. There are, however, some bi-directional pipelines which
enable flow in both directions. These pipelines can be useful when oversupply of one or more
Natural gas has emerged as the fuel of choice for various uses. It is most extensively used by
domestic consumers cooking and heating, petrochemical and fertilizer industries as both fuel and
feedstock, and by power producers for generating electricity. While there are substitutes for
63
natural gas in industry (e.g. naphtha for fertilizer plants), current prices makes natural gas more
competitive Worldwide trade in natural gas is also growing rapidly, and it reserves are more
distributed throughout the world than those of oil. The proven reserves have more than doubled
in the last 20 years, increasing the world R/P to 64.1 years in 1997 (BP, 1998). The following are
ii. About ninety-seven percent (97%) of the natural gas consumed in Ghana is produced in
Nigeria.
iii. Ghana could soon become part of the larger global market by increasing imports of LNG
and also become an LNG exporter later after its oil discovery.
iv. West African gas pipeline network will reduce or end the impact of harmful gas flaring
We used the Prim minimal spanning tree algorithm for directed graph G= (V, E], where V is the
set of nodes and E is the set of edges. Each edge is defined by a node pair (h, k), each edge has a
non-negative length. We start at a given node and sequentially add an edge to tree at each of the
iterations.
The weight of the graph is the total distance covered in the regional capitals.
64
V is a set of vertices (nodes)
Each edge is a pair of vertices. In other words, each element of E is a pair of elements of
V.
X, Y, Z.}
Edges (E) of map= pair (h, k); where h= A, B, C,…, Z; and k= A, B, C, ..., Z.
The algorithm continuously increases the size of a tree, one edge at a time, starting with a tree
Input: A non- empty connected weighted graph with vertices V and edges E in which the
Initialize: Vnew = {A}, where A is an arbitrary node (starting point) from V, E new = { }
o Choose an edge (h, k) with minimal weight such that h is in V new and k is not (if
there are multiple edges with the same weight, any of them may be picked)
Minimum spanning tree: tree that connects all the vertices and minimizes
65
Since the connection problem is a network problem, it can be solved by the shortest path method.
But because of its special nature, it can be solved more easily by Prim‟s algorithm, which is
ALGORITHM)
Dijkstra‟s algorithm, conceived by Dutch computer scientist Edsger Dijkstra in 1959, is a graph
search algorithm that solves the single-source shortest path problem from a graph with
nonnegative edge path costs, producing a shortest path tree. This algorithm is often used in
For a given source vertex (node) in the graph, the algorithm finds the path with the lowest cost
(i.e. the shortest path) between that vertex and that of every vertex. It can also be used for finding
costs of shortest paths from a single destination vertex by stopping the algorithm once the
shortest path to the destination vertex has been determined. For example, if the vertices of the
graph represent cities and edge path costs represent driving distances between two pairs of cities
connected by a direct road, Dijkstra‟s algorithm can be used to find the shortest route between
one city and all other cities. As a result, the shortest path first is widely used in network routing
protocols, most notably IS-SI and OSPF (Open Shortest Path First).
3.7.2 ALGORITHM
Let the node we are starting be called the initial node. Let a distance of the node Y be the
distance from the initial node to it. Dijkstra‟s algorithm will assign more initial distance values
66
(i) Assign to every node the distance value. Set it to zero for our initial node and infinity
(iii)For current node, consider all its as unvisited neighbours and calculate their distance
(from the initial node). For example, if current node (A) has distance of 8, and an edge
connecting it with another node (B) is 2, the distance to B through A will be 8+2=10.
If the distance is less than the previously recorded distance (infinity in the beginning,
(iv) When we are done considering all neighbours of the current node, mark it as unvisited.
A visited node will not be checked ever again; its distance now recorded is final and
minimal.
(v) Set the unvisited node with the smallest distance (from the initial node) as the next
Dijkstra‟s algorithm finds the shortest path from the source node S to all other node in a network
with nonnegative arc lengths. Dijkstra‟s algorithm maintains a distance label d (i) with each node
i, which is an upper bound on the shortest path length from the source node to each node i. at any
intermediate step, the algorithm divides the nodes of the network under considerations into two
groups: those which it designates as permanently labeled (or permanent) and those which from it
designates are temporarily labeled (or temporal). The distance label to any permanent node
represents the shortest distance from the source node to that node.
The basic idea of the algorithm is to find out from the source node S and permanently labeled
nodes in the order of their distance from the node S. initially, node S is assigned a permanent
label of zero, and each other; node j a temporary label equally to infinity. At each iteration, the
label of a node I is its shortest distance from the source node along the path whose internal node
67
(i.e. nodes other than S or I itself) are permanently labeled. The algorithm selects a node i with
minimum temporary label (breaking ties arbitrarily), makes it permanent, and reaches out from
node- that is, scans all edges/arcs emanating from node I to update the distance labels of adjacent
nodes. The algorithm terminates when it has designated all nodes permanent.
Step 2: assign temporary labels to all the vertices that are connected directly to the most recently
Step 3: choose the vertex with the smallest temporary label and assign a permanent label to that
vertex.
Step 4; Repeat step 2 and 3 until all vertices have been permanent labels.
Step 5; find the shortest path by tracing back through the network.
Note: Recording the order in which we assign permanent labels to the vertices is an essential part
of the algorithm.
68
3.7.3 ALL-PAIRS SHORTEST PATH PROBLEM
The shortest path between two nodes might not be a direct edge between them, but instead
involve a detour through other nodes. The all-pairs shortest path problem requires that we
determine shortest path distance between every pair of nodes in the network.
The Floyd- Warshall algorithm obtains a matrix of shortest path distances within O {n3}
Let dk(i, j) represents the length of the shortest path from node i to node j subject to the condition
that this path uses the node 1, 2, 3,…,k-1 as internal nodes clearly, dn+1 (i, j) represents the actual
shortest path distance from i to j. the algorithm first computes d1(i, j) for all node pairs i and j.
using d1 (i, j), it then computes d2 (i, j) for all pairs of nodes i and j. It repeats the process until it
obtains dn+1 (i, j) for all node pairs i and j when it terminates. Given d k (i, j), the algorithm
computes dk+1 (i, j) = min {dk (i, k), dk (i, j)}. The Floyd-Warshall algorithm remains of interest
3.7.5 MINIMUM-CONNECTOR
2. A spanning tree is a sub graph that includes all the vertices in the original graph is also a
69
These are all spanning tree.
B D B
A D F
C E
F
A C E
B D
A F
C E
Prim‟s algorithm solves the shortest path problem for a directed graph with non-negative edge
weights. For example, if the vertices of the graph represents gas well or roads connecting the
towns and edge weights represent distances between pair of wells (or towns ) connected by a
pipeline, this algorithm can be used to find the shortest route between two gas wells (or towns).
The algorithm was discovered in 1930 by mathematician Vojtech Jarnik and later independently
by computer scientist Robert Prim in 1957 and rediscovered by Dijkstra in 1959. Therefore it is
sometimes DJP algorithm or Jarnik algorithm. The algorithm builds a spanning tree from scratch
by fanning out from a single node and adding arc one at a time. It means a tree spanning on a
subset S of nodes and a nearest neighbour to S. the algorithm does so by identifying an arc (i, j)
of minimum cost in the cut [S, S]. It adds arc (i, j) to the tree, node j to S repeats this basic steps
70
until all nodes have been connected. This algorithm is said to be “greedy” since it picks the
immediate best option available without taking into account the long-term effects of the choices
Step 3; join on the vertex which is nearest to either of those already connected.
Step 5; add the lengths of all the edges included in the minimum length spanning tree.
The correctness of the algorithm follows directly from the fact that each arc that is added to the
tree is contained in some minimum spanning tree with the arcs that have been selected in the
previous step. The matrix formulation of Prim‟s algorithm implemented in this thesis is given
below.
Step 1; choose a start vertex and delete all elements in that vertex‟s row and arrow its column.
Step 2; Neglecting all deleted terms, scan all arrowed columns for the lowest element and circle
that element.
Step 3; Delete the circled element‟s row and arrow its column.
Step 6; write down the arcs in the order you selected them (Amponsah, 2007).
71
3.7.7 A MINIMUM CONNECTOR PROBLEM
A company is installing a system of cables to connect all towns in the region. The lower case
variables in the network show distances in kilometers. The question is what is the least amount
of cable needed?
j
h
q
D E
g n
p
k
A m B i C
Prim‟s algorithms works from a start point and builds up the spanning tree step by step,
connecting edges into the existing solution. It can be applied directly to the distance matrix, as
well as to work the network itself, so it is more suitable for using a computer if the network is
large.
Step 2; join this vertex to the nearest vertex directly connected to it.
72
Step 3; Join the next nearest vertex, not already in the solution to any vertex in the solution,
Maps and field work are required for pipeline routine, pipeline design and construction. The
information and data needed for the analysis will be gathered from the Ghana Highways
3.9 SUMMARY
In this chapter, we considered the method and the modeling of the gas pipeline problem into
The next chapter will focus on analyzing the gas pipeline path from Takoradi to all the regional
capital towns and the environmental value and damages revealed by the gas pipeline network.
73
CHAPTER 4
4.0 INTRODUCTION
This chapter analyses the problem of designing the pipeline networks. The existing roads linking
the regional capital towns are being used to provide access to the construction of the pipelines
connection. The data involves length of road distances (km) between the various regional capital
towns. The methodology outlined in the chapter three will be used to identify the overall network
layout from Takoradi to all the regional capital towns, the author‟s own optimal layout using
Prim‟s algorithm and the environmental damages. For easy computation, we let
74
4.1 COMPUTATIONAL PROCEDURE AND ANALYSIS OF DATA
J
275
314 161
375
567
382
G
130 504
436 336
200
305 272 135
E
222 D
85
166
C
145
A 84 B
Figure 4.1
Pipeline Network (in km) from Takoradi to all the regional capital towns in
Ghana
75
Source: Ghana High Authority (Not drawn to scale)
Total distance covered by the pipe network in the above map is 5,094km
The pipelines used in these network systems can measure anywhere from 6 to 48 inches in
diameter, although certain component pipe section can consist of small-diameter pipe that is as
small as 0.5 inch in diameter. However, this small-diameter pipe is usually used only in
gathering and distribution systems, although some is used for control-line or gauge-line
purposes. Mainline pipes, the principal pipeline in a given system, are usually between 16 to 48
inches in diameter. Lateral pipelines, which deliver natural gas to or from the mainline, are
typically between 6 and 16 inches in diameter. Most major interstate pipelines are between 24
76
Table 4.1: DISTANCE MATRIX OF THE NETWORK
A B C D E F G H I J
A ∞ 84 ∞ ∞ ∞ 105 436 ∞ ∞ ∞
B 84 ∞ 145 ∞ ∞ 222 ∞ ∞ ∞ ∞
E ∞ ∞ 85 135 ∞ 200 ∞ ∞ ∞ ∞
Prim‟s Algorithm is now applied to the distance matrix above with the solution taken straight
from the graph shown alongside.
77
MATRIX METHOD
Choose the starting vertex, A. Delete the row A. Look for the smallest entry in column A.
Table 4.2:Information1
A B C D E F G H I J
A ∞ 84 ∞ ∞ ∞ 105 436 ∞ ∞ ∞
B 84
84 ∞ 145 ∞ ∞ 222 ∞ ∞ ∞ ∞
E ∞ ∞ 85 135 ∞ 200 ∞ ∞ ∞ ∞
78
AB is the smallest edge joining A to the other vertices. Put edge AB into solution. Delete row B.
Look for the smallest entry in columns A and B.
1 2
A B C D E F G H I J
B 84 ∞ 145 ∞ ∞ 222 ∞ ∞ ∞ ∞
C ∞ 145
145 ∞ 166 85 272 ∞ ∞ ∞ ∞
E ∞ ∞ 85 135 ∞ 200 ∞ ∞ ∞ ∞
84 B
79
BC is the smallest edge joining A, B and C to the other vertices. Put edge BC into solution.
Delete row C. Look for the smallest entry in columns A, B and C.
1 2 3
A B C D E F G H I J
E ∞ ∞ 85
85 135 ∞ 200 ∞ ∞ ∞ ∞
145
B
84
A
80
CE is the smallest edge joining A, B, C and E to the other vertices. Put edge CE into solution.
Delete row E. Look for the smallest entry in columns A, B, C and E.
1 2 3 4
A B C D E F G H I J
D ∞ ∞ 166 ∞ 135
135 336 ∞ ∞ 504 567
E ∞ ∞ 85 135 ∞ 200 ∞ ∞ ∞ ∞
85
C
Figure 4.5: solution 4
145
B
84
A
81
ED is the smallest edge joining A, B, C, E and D to the other vertices. Put edge ED into solution.
Delete row D. Look for the smallest entry in columns A, B, C, E and D
1 2 3 5 4
A B C D E F G H I J
135
85
Figure4.6: Solution 5
C
145
B
82
84
A
DF is the smallest edge joining A, B, C, E, D and F to the other vertices. Put edge DF into
solution. Delete row F. Look for the smallest entry in columns A, B, C, E, D and F
1 2 3 5 4 6
A B C D E F G H I J
G 436 ∞ ∞ ∞ ∞ 130
130 ∞ 375 ∞ ∞
135
F
E
200
85
145
B
84
A
83
FG is the smallest edge joining A, B, C, E, D, F and G to the other vertices. Put edge FG into
solution. Delete row G. Look for the smallest entry in columns A, B, C, E, D, F and G
1 2 3 5 4 6 7
A B C D E F G H I J
H ∞ ∞ ∞ ∞ ∞ ∞ 375
375 ∞ 314 275
G D
130 135
F
E
200
85
145
B
84
A
84
GH is the smallest edge joining A, B, C, E, D, F, G and H to the other vertices. Put edge GH into
solution. Delete row H. Look for the smallest entry in columns A, B, C, E, D, F, G and H.
1 2 3 5 4 6 7 8
A B C D E F G H I J
J ∞ ∞ ∞ 567 ∞ ∞ ∞ 275
275 161 ∞
375
G D
130
135
F
E
200
85
Figure 4.9: Solution 8
145
B
84 85
A
HJ is the smallest edge joining A, B, C, E, D, F, G, H and J to the other vertices. Put edge HJ
into solution. Delete row J. Look for the smallest entry in columns A, B, C, E, D, F, G, H and J.
1 2 3 5 4 6 7 8 10 9
A B C D E F G H I J
J
275
375
G
D
130
135
F
E
200
145
B 86
84
A
87
JI is the smallest edge joining A, B, C, E, D, F, G, H, J and I to the other vertices. Put edge JI
into solution.
1 2 3 5 4 6 7 8 10 9
A B C D E F G H I J
275
I 161
375
G D
130
135
F
E
200
145
B
84 88
A
Table 4.12: Summary of Matrix Representation of the Pipeline Network.
A B C D E F G H I J
A ∞ 84 ∞ ∞ ∞ ∞ ∞ ∞ ∞ ∞
B 84 ∞ 145 ∞ ∞ ∞ ∞ ∞ ∞ ∞
C ∞ 145 ∞ ∞ 85 ∞ ∞ ∞ ∞ ∞
D ∞ ∞ ∞ ∞ 135 ∞ ∞ ∞ ∞ ∞
E ∞ ∞ 85 135 ∞ 200 ∞ ∞ ∞ ∞
F ∞ ∞ ∞ ∞ 200 ∞ 130 ∞ ∞ ∞
G ∞ ∞ ∞ ∞ ∞ 130 ∞ 375 ∞ ∞
H ∞ ∞ ∞ ∞ ∞ ∞ 375 ∞ ∞ 275
I ∞ ∞ ∞ ∞ ∞ ∞ ∞ ∞ ∞ 161
J ∞ ∞ ∞ ∞ ∞ ∞ ∞ 275 161 ∞
89
4.2 DISCUSSION
Figure 4.1 represents the pipe network to the regional capital towns in Ghana.
The alphabets represent nodes at the regional capital towns. The above map was designed out of
From the table, distances between nodes are indicated and nodes without direct contact are given
infinity sign. The total distance covered by the map is computed as 5094km.
Using the Prim‟s algorithm, the optimal solution and the minimized connection distance are
drawn below.
90
J
275
161
H
375
G D
130
135
F
200
E
85
C
145
B
84
A
91
4.3 ANALYSIS
We observe the map has become smaller from the original map.
The number of pipes in the network has also reduced as compare to that of the original
The difference between the distance covered by the old map and the distance covered by
the new map after the use of the Prim‟s algorithm is computed as:
The percentage reduction in the distance as compared to the distance of the original map
is computed as 3504/5094 × 100 = 68.79%. This means that after the use of the Prim‟s
algorithm, there is 68.79% reduction in the pipe network of the original map.
The application of the findings is that, the pipe network connecting the regional capital
Supposing a pipeline with a specified diameter is GH¢x for y metre long, then the price
The difference between the total price of the original map and that of the new map after
92
GH¢5094x/y – GH¢1590/y = GH¢3504x/y
The environmental damage potentially caused by pipelines include disturbance of soil, forest,
rivers and wetlands, noise from compressor stations to during pipeline discharge, and risk of
pipeline leakage or rapture. Environmental valuation is a method by which monetary values are
placed upon environmental attributes which enables these attributes and associated damage or
Placing any kind of value on the environment is a difficult thing to do. This is true with the
almost all “public good”. A public good is something to which everyone has not paid for in the
same fashion as private goods, such as automobiles and microwave ovens. Clean air, beautiful
views and quietness may all be considered to be public goods. The concept discussed in the
chapter have been extensively applied in recent years among researchers and academics looking
for new ways to approach environmental problems, and this concept are discussed in detail in
Dott et al (1996b)
Their applicability in industry has been limited because of practical difficulties associated with
the same of the valuation methods. In addition, researchers have found widely varying results for
different studies, making the ‟correct‟ correct value difficult for determine.
Standard cost-benefit analysis weighs the cost and benefits of a proposed project, almost always
in monetary terms. Usually, the project with the highest Net Presents Value (NPV), or time-
93
adjusted net benefit, is the one which is pursued. Using valuation techniques, environmental cost
can be added into the cost-benefit analysis, forming one of the project costs.
The environmental costs could be used to compare the economic viability of two or more
alternative projects, the shortest link-length pipeline network connecting a number of nodes to a
gas processing plant could represent the cost optimal pipeline network if environmental costs
The cost of constructing the pipeline, and the cost associated with the environmental damage
incurred are included using the approach. On the other hand, some of the links around
environmentally sensitive area (ESA‟s), such as wetlands or river crossings are re-directed
resulting in a longer total pipeline length. In this case, the construction cost of the longer pipeline
The two options described above would not entirely avoid environmental degradation, but the
comparison of the two will indicate the preferred route. If society values wetland and rivers, then
As expressed by Zube (1980) information on the value of the can also be helpful for
governmental agencies in setting policies and direction for spending on our environmental
property zoning and development planning, to name a few areas. The difficulty people may have
placing monetary value on environmental goods and human life are catered for by the engineers
who understand that some risk of death or injury associated with their designs, as much as we try
to minimize them.
94
4.4.2 AGRICULTURE AND SOILS
The majority of the proposed pipeline route will crosses agricultural land. Much of this is in
intensive arable use, producing crops and root crops. There will inevitably be some temporally
All migration measure will be detailed within formal legal agreements with landowners and
occupiers. Soil will be handled in accordance with best practice guidelines to protect soil
fertility. Pre-and post construction remedial land drainage system will be installed to ensure
agricultural drainage system continue to function. All reasonable precautions in line with current
best practice will be taken to prevent the spread of plant and animal and agricultural weeds
during construction. Arable land is expected to be returned to normal use by the following crop
growing season.
During the construction phase, there is the potential for sediments or pollutants to be released to
water courses or ground water as a result of surface run off or leaching. Where open cut crossing
are proposed, there is the potential for physical damage to watercourse. All construction will be
taken in accordance with EPA guidance to ensure that adverse impacts on the water environment
are avoided. Method statements will be prepared and agreed with the EPA and
landowner/occupiers for all work at watercourse crossing (including reinstatement). The decision
to employ trenchless techniques for major watercourse crossing will prevent disturbances to the
river bed and flood defenses and limit the potential for pollution of these features. The proposed
pipeline will not reduce flood plain storage capacity as it will be buried.
95
4.4.4 ECOLOGY
Arable land is generally of limited ecological value. However, a number of protected species
occur along the proposed pipeline route. There will be temporary loss of habitats during
The proposed pipeline route and construction techniques should be selected to minimize
4.4.5 EMISSIONS
These potential impacts will be mitigated and controlled through a construction Environmental
Management Plan and consultation with the local Environmental Health Officer.
Construction of the proposed pipeline will result in temporary increase in traffic flow with
potential impacts to offer road users. The greatest transportation requirement during construction
is the delivery of the steel pipe lengths to the working width. Construction of the road crossings
Traffic Management Plan (TMP) to be agreed with the Ghana Highways Authority (GHA) prior
96
The TMP will identify any road restrictions and agreed traffic routes, specify signage and
temporary traffic controls and any additional measures to minimize impacts. The number of
journeys will be minimized and materials could be sourced locally where feasible. The proposed
pipeline will be installed under roads using trenchless construction techniques to minimize traffic
construction. Access will be maintained while the pipe is being installed by placing across trench
or temporary diversions.
4.4.7 SOCIO-ECONOMICS
The will be some economic benefit during pipeline construction as a result of local employment
opportunities and the use of local services and businesses, although this is not yet considered to
be significant at a regional scale. There will also be a negative and economic impact as a result
scale.
In addition, there may be some temporary loss of tranquility and visual attractiveness of the area
local to the pipeline and disruption to Right of Way during pipeline construction.
Measures including dust suppression and noise reduction techniques will be taken to reduce the
impact of the construction works on the amenity value of roads and Public Right of Way.
Rights of Way will generally be kept open, during construction by incorporating gates or stiles
along the working width. Where it is necessary to temporarily close or divert Public Rights of
Way to ensure public safety, this will be agreed by the parties involved.
97
4.4.8 GEOLOGY AND GROUND CONDITIONS
There is the potential for the pollution of watercourses and drift deposits (shallow geology) as a
result of accidental fuel, oil or waste leaks or from importation of contaminated materials. A
method statement for dealing with unexpected contamination will be prepared and submitted to
the Authority in charge for agreement prior to any works commencing on site. All secondary
materials imported to the site will have previously been analyzed for the presence of leachable
contaminants.
98
CHAPTER 5
5.0 INTRODUCTION
This chapter comprises drawing of conclusions based on the outcome of the experimentations.
5.1 CONCLUSION
The findings in chapter 4 signify that Prim‟s Algorithm can be used to reduce the network
systems. The supply of natural gas from Takoradi to other parts has become very difficult due to
no pipe line network systems connecting the gas from Takoradi to other regional capital towns in
The total distance covered by the pipe line network if the existing road networks are used from
Takoradi to all the regional capitals towns in Ghana is 5,094km. After Prim‟s Algorithm was
used, the total distance covered is 1,590km. Considering the original map, total distance covered
by the pipe network system was reduced by 68.79% when the Prim‟s Algorithm was used to
Environmental damages are other significant findings in this research. Implementations of these
environmental solutions in this research will reduce the impact of these problems in construction
of the pipe network. In as much as some environmental corridors are deemed impossible for the
disinclined to implement this Algorithm when the environment permit (Donkoh, 2010)
In conclusion, the objectives of the research have been achieved by minimizing the pipe network
and providing environmental solutions. It also means that the Prim‟s Algorithm can be used in
99
reducing all pipe networks in the country to smaller, more efficient and less costly in the country.
To minimize pipe network systems, the Prim‟s Algorithm should be used to reduce the distance
coverage and the price involved in the changing of the pipe network systems.
5.2 RECOMMENDATIONS
The government in need of providing gas pipe network should employ operational
Existing roads with many curves should be straightened to further minimize the distances
Communication, transportations and network can also employ the use of shortest path
problem especially, Prim‟s Algorithm to plan their network systems. It will minimize the
network, save money and human resources but still meet the need of their customers.
The author advises that the optimal network flow of the gas pipeline should follow the
Individuals can also use the Prim‟s Algorithm to plan their pipe network, electrical
Environmental factors should be considered when pipe networks are being layout.
100
REFERENCES
1. Ahuja. R. K., Magnanti L.T., Orlin J. B and Reddy R.M. (1992). Applications for Network
Optimization.
2. Ahuja. R. K., Magnanti L.T. and Orlin J. B (1989). Some recent advances in network flow.
3. Amiteye, B.T. (2002). Distribution and Ecology of Nesting Sea Turtles in Ghana.
5. Arogundade O.T and Akinwale A.T (2009). Application of Prim‟s Algorithm to a Profit
6. Bern, M.W. and R.L. Graham (1989). The shortest network problem. Scientific American
Journal, pp 84-89.
Canada.
9. Brimberg. J., Mladenović N. (2003). Oil pipeline design problem. Operation Research, Vol.
10. Chang (2008).The reason why it is beneficial to combine with graph theory and board game.
12. Divoky,(1990).Framework for solving shortest-path problem with positive edge weights.
13. Donkoh, E. K. (2010) Optimal Network Flow; the West African Gas Pipeline Project.
101
14. Dott, D.R. putting the environment into NPV calculation-Contingent valuation to qualify
pipeline environmental costs. Proceedings of the 1st International Pipeline Conference, Vol.
2, pp. 1271-1277
15. George, R. and Mortenson.P, (1995). Toward Continental Gas Market: The Integration of
17. Goldberg (2006).Proposed shortest path algorithms that use a search in combination with a
new graph-theoretic lower bounding technique based on landmarks and triangle inequality.
18. Gonnia (2007).Parallel implementation of Prim‟s algorithm for finding a minimum spanning
19. Hwee, M. Lin, R., Poe, W. A, (2007).Primary Reformer Operation :A Unique Application of
20. Houardy,(2010). The Floyd Warshall Algorithm is a simple and widely used algorithm to
compute shortest paths between all pairs of vertices in an edged weighted directed graph.
21. Lysgaard,(2000). New algorithm for finding the shortest path from a source to a single sink
in a network.
22. Li. C. and Maini, P.K. (2005). An evolving network model with community structure.
24. McCarthy (2009).The application of two well known graph algorithms; Edmonds ‟algorithm
25. Martel (2002).The expected performance of Prim‟s minimum spanning tree (MST) algorithm
26. Melzak, Z. A (1961). On the problem of Stenier. Canadian Mathematical Bulletin, Vol. 4/2e,
pp. 143-148
102
27. Manci, K.M, Gladwin, D.N., Villella, R. and Cavendish M.G, (1988). Effects of Aircraft
Noise and Sonic Booms on Domestic Animals and Wildlife: A literature Synthesis. U.S. Fish
and Wildlife Service. National Ecology Research Center, Ft. Collins, CO NERC-88/29. pp.
8.
28. Miehle, W. (1958). Link-length minimization in networks. Operations Research Vol. 6/2e,
pp. 232-243.
29. Nesbitt, D.M (1998). Methodology of the GRI North Regional Gas Supply-Demand Model.
30. Owusu Frank (2010).Minimum connection of pipe network system: Ghana Water Company
32. Pekal J. Steinhaus H. (1950). Minimum Spanning Tree with Application to Anthropology,
33. Rabbani, (2008). Distribution network design problem in a multi-product supply chain
system.
matrix.
35. Rothfard. B., Frank. K. (1970). Optimal Design of Offshore Natural-Gas Pipeline Systems.
36. Prim R.C., Shortest connection networks and some generalizations. In: Bell System technical
computer science.
38. Sorrel, F.Y. (1982). Oil and Gas Pipelines in Coastal North Carolina: Impacts and Routing
40. US Department of Energy Outlook. (2005). Oil and Gas Journal, USA.
43. Wang, (2005).Multiple pairs of shortest path problem (MPSP) arise in many applications.
44. Wearmouth, S.A. (1994) Utilization of Gas in Alberta-Canada. Master‟s Thesis, University
104
APPENDIX A
Branch Name Start Node End Node General Cost Pipe Length General Cost
A 0 1 0 y 0
B 1 2 84x y 84x
C 2 3 145x y 145x
E 3 4 85x y 85x
D 4 5 135x y 135x
F 5 6 200x y 200x
G 6 7 130x y 130x
H 7 8 375x y 375x
I 8 9 275x y 275x
J 9 10 161x y 161x
Total 1590x
105
APPENDIX B
A 0 1 0 0
B 1 2 84x 84x
C 2 3 145x 229x
E 3 4 85x 314x
D 4 5 135x 449x
F 5 6 200x 649x
G 6 7 130x 779x
H 7 8 375x 1154x
I 8 9 275x 1429x
J 9 10 161x 1590x
106